The Cannabinoids Δ⁸THC, CBD, and HU-308 Act via Distinct Receptors to Reduce Corneal Pain and Inflammation

Abstract

Background and Purpose: Corneal injury can result in dysfunction of corneal nociceptive signaling and corneal sensitization. Activation of the endocannabinoid system has been reported to be analgesic and anti-inflammatory. The purpose of this research was to investigate the antinociceptive and anti-inflammatory effects of cannabinoids with reported actions at cannabinoid 1 (CB₁R) and cannabinoid 2 (CB₂R) receptors and/or noncannabinoid receptors in an experimental model of corneal hyperalgesia.

Methods: Corneal hyperalgesia (increased pain response) was generated using chemical cauterization of the corneal epithelium in wild-type (WT) and CB₂R knockout (CB₂R^−/−) mice. Cauterized eyes were treated topically with the phytocannabinoids Δ⁸-tetrahydrocannabinol (Δ⁸THC) or cannabidiol (CBD), or the CBD derivative HU-308, in the presence or absence of the CB₁R antagonist AM251 (2.0 mg/kg i.p.), or the 5-HT_1A receptor antagonist WAY100635 (1 mg/kg i.p.). Behavioral pain responses to a topical capsaicin challenge at 6 h postinjury were quantified from video recordings. Mice were euthanized at 6 and 12 h postcorneal injury for immunohistochemical analysis to quantify corneal neutrophil infiltration.

Results: Corneal cauterization resulted in hyperalgesia to capsaicin at 6 h postinjury compared to sham control eyes. Neutrophil infiltration, indicative of inflammation, was apparent at 6 and 12 h postinjury in WT mice. Application of Δ⁸THC, CBD, and HU-308 reduced the pain score and neutrophil infiltration in WT mice. The antinociceptive and anti-inflammatory actions of Δ⁸THC, but not CBD, were blocked by the CB₁R antagonist AM251, but were still apparent, for both cannabinoids, in CB₂R^−/− mice. However, the antinociceptive and anti-inflammatory actions of HU-308 were absent in the CB₂R^−/− mice. The antinociceptive and anti-inflammatory effects of CBD were blocked by the 5-HT_1A antagonist WAY100635.

Conclusion: Topical cannabinoids reduce corneal hyperalgesia and inflammation. The antinociceptive and anti-inflammatory effects of Δ⁸THC are mediated primarily via CB₁R, whereas that of the cannabinoids CBD and HU-308, involve activation of 5-HT_1A receptors and CB₂Rs, respectively. Cannabinoids could be a novel clinical therapy for corneal pain and inflammation resulting from ocular surface injury.

Introduction

The cornea is a thin, transparent dome-shaped avascular tissue that is densely innervated by sensory nerve endings.^1,2 Damage to these nerve endings, resulting from surgery, trauma, neurological disease, or infection, may develop into corneal neuropathic pain (CNP).² CNP is a clinically significant problem characterized by persistent hyperalgesia, debilitating pain, photoallodynia, burning, stinging, dryness, and inflammation.³ Corneal damage can also result in an inflammatory response that involves the production of proinflammatory cytokines, neovascularization, recruitment of leukocytes, and release of neuropeptides producing inflammatory pain.^4,5

Existing pharmacotherapies for ocular pain, inflammation, and CNP include topical corticosteroids, tricyclic antidepressants, GABAergic drugs, and opioids.^3,6 These treatments, however, frequently fail to provide adequate pain relief and are associated with side effects.^3,6 Therefore, new therapies that can alleviate pain and symptoms associated with CNP have fewer side effects and can resolve corneal inflammation are urgently required. One drug target that may have a role in the modulation of pain and inflammation is the endocannabinoid system (ECS).^7,8

The ECS is an endogenous lipid signaling system that includes two G-protein-coupled receptors, cannabinoid 1 receptor (CB₁R) and cannabinoid 2 receptor (CB₂R), endocannabinoids, and cognate enzymes for biosynthesis and degradation of endocannabinoids.^9,10 CB₁Rs are widely expressed in many tissues, including in the central and peripheral nervous systems, where activation of CB₁R modulates neurotransmitter release, and nerve activity.^11,12 CB₂R is highly expressed on immune cells and its activation is anti-inflammatory, resulting in decreased production of proinflammatory mediators and a reduction in leukocyte recruitment.^13–15 Drugs that enhance activation of the ECS, including activation of both CB₁R and CB₂R, have shown efficacy in experimental models of pain and inflammation, including neuropathic pain.^7,9,16,17

Cannabinoids have not been extensively studied in ocular surface pain and inflammation; however, CB₁R is expressed in the corneal epithelium and endothelium in rodents and primates,¹⁸ and activation of CB₁R has been reported to inhibit neuropeptide-induced sensitization of transient receptor potential cation channel subfamily V member 1 (TRPV1) in afferent neurons.¹¹ Under nonpathological conditions, CB₂R expression is low in the cornea and anterior ocular structures; however, increased CB₂R expression in anterior ocular tissues has been suggested in experimental uveitis, where CB₂R activation produces anti-inflammatory effects.^19–21 Taken together, these studies suggest that cannabinoids that activate CB₁R and/or CB₂R may be useful for mitigating corneal pain and inflammation.

In this study, we used a mouse model of corneal hyperalgesia to investigate the antinociceptive and anti-inflammatory effects of several cannabinoids that act at CB₁R and/or CB₂R, or noncannabinoid receptors. These included Δ⁸-tetrahydrocannabinol (Δ⁸THC), a more stable isomer of Δ⁹-tetrahydrocannabinol (Δ⁹THC), cannabidiol (CBD), and the CBD derivative HU-308. Both Δ⁸THC and Δ⁹THC produce antinociceptive effects in pre-clinical models with similar potency via activation of CB₁R.^22–26 CBD lacks the behavioral effects of THC at CB₁R, and may produce pharmacological actions through the activation of noncannabinoid receptors.^27–29 HU-308 is a selective and highly potent agonist at CB₂R,³⁰ and has previously been shown to reduce lipopolysaccharide-induced intraocular inflammation.¹⁹

Materials and Methods

Experimental animals and corneal injury model

All animal care and experimental procedures complied with the Canadian Council for Animal Care guidelines (www.ccac.ca/) and were approved by the Dalhousie University Committee on Laboratory Animals. Male BALB/c (20–30 g; Charles River Laboratories International, Inc., Wilmington, MA) and CB₂R knockout mice (CB₂R^−/−) were used for experiments. CB₂R^−/− mice were obtained by crossing male C57BL/6J CB₂R^−/− mice (strain B6.129P2-Cnr2tm1Dgen/J; Jackson Laboratory, Bar Harbor, ME) with inbred BALB/c female mice (Charles River) for 10 generations. Genetic loss of CB₂R (Cnr2) was confirmed via polymerase chain reaction genotyping using DNA extracted from ear punches with an Accustart II Mouse Genotyping Kit (Quanta Biosciences, Gaithersburg, MD) according to the manufacturer's instructions. Primer sequences were as follows: mouse CB₂ mutant forward (moIMR0086) 5′- GGG GAT CGA TCC GTC CTG TAA GTC T-3′; mouse CB₂ wild-type (WT) forward (oIMR7552) 5′- GGA GTT CAA CCC CAT GAA GGA GTA C-3′; mouse CB₂ common reverse (oIMR7552) 5′- GAC TAG AGC TTT GTA GGT AGG CGG G−3′ with a single product at ∼550 bp for CB₂R^−/−, a single product at ∼385 bp for WT, and two products at ∼550 and 385 bp for heterozygous mice. Mice were housed in groups of 3–5, kept on a light/dark cycle (07:00–19:00/19:00–07:00), and fed ad libitum.

Corneal injury was induced using a protocol adapted from a model of corneal hyperalgesia previously described in rats by Wenk and Honda.³¹ Briefly, mice were anesthetized using 2–3% isoflurane gas. The center of the cornea on both eyes was cauterized with silver nitrate (MedPro^®, 75% silver nitrate, 25% potassium nitrate; AMG Medical Inc., Montreal, QC, Canada) using a micro-applicator brush (Centrix, Inc., Shelton, CT). The micro-applicator brush was held in contact with the cornea for 2 sec, producing a distinct superficial white lesion of 1 mm in diameter, injuring the epithelial cell layer only. The cauterized eyes were then rinsed with saline and an ocular lubricant (Systane^®; Alcon Canada, Inc., Dorval, QC, Canada) was applied to reduce corneal drying. Mice recovered fully from anesthesia within 3–5 min postcauterization. Mice were euthanized at 6 or 12 h postcauterization, and the eyes were enucleated for immunohistochemical analysis.

Assessment of behavioral pain sensitization

At 6 or 12 h postcauterization, 5 μL of 1 μM capsaicin was applied topically to the cauterized eyes to elicit a pain response.³² A sham control group was induced by touching the micro-applicator brush to the cornea for 2 sec in the absence of sliver nitrate, keeping all other parameters the same. Immediately following the application of a single dose of capsaicin, the behavioral response was video recorded for 30 sec. Videos were analyzed offline in slow motion, where the number of blinks, squints, and eye wipes to capsaicin challenge was summed to give a pain score.

Immunohistochemistry

At 12 h following corneal cauterization, eyes were enucleated and fixed in 4% PFA, followed by 30% sucrose overnight. Corneal sections (12 μm) were cut using a Leica CM1850 cryostat (Wetzlar, Germany). Sections were washed in phosphate-buffered saline (PBS) and blocked for nonspecific binding (10% normal goat serum in 0.5% Triton-X/PBS; Sigma-Aldrich, Oakville, ON, Canada) for 2 h, followed by a 48-h incubation in purified rat anti-Ly-6G antibody (1:200; Abcam, Cambridge, MA). Ly-6G is a glycosylphosphatidylinositol-anchor protein expressed predominantly on neutrophils; the anti-Ly-6G antibody allows for detection of these cells.³³ Sections were then washed with PBS and incubated with a secondary antibody (1:500, goat anti-rat Alexa Fluor^® 488, Jackson ImmunoResearch Laboratories, Inc., West Grove, PA). Stained sections were washed in PBS and mounted on Superfrost slides (Fisher Scientific, ON, Canada) using Fluoromount (Sigma-Aldrich).

Neutrophil migration, indicative of an innate immune response, was quantified in corneal sections at 20×magnification using a Zeiss Axiovert 200 M microscope (Zeiss, Thornwood, NY). Three representative images were taken from each section of the right and left corneal peripheries and from the center of the cornea, respectively. The total number of neutrophils from these three images was counted for each section, and summed to represent the total neutrophil number for a single corneal section. A total of 5–12 sections with 120 μM intervals from each eye were analyzed and the neutrophil number was averaged.

Drugs and solutions

Δ⁸THC ([6aR,10aR]-6,6,9-trimethyl-3-pentyl-6a,7,8,10a-tetrahydrobenzo[c]chromen-1-ol; Cayman Chemical, Ann Arbor, MI), CBD (2-[(1R,6R)-3-Methyl-6-(1-methylethenyl)-2-cyclohexen-1-yl]-5-pentyl-1,3-benzenediol; Cayman Chemical), HU-308 (4-[4-(1,1-Dimethylheptyl)-2,6-dimethoxyphenyl]-6,6-dimethylbicyclo[3.1.1]hept-2-ene-2-methanol; Tocris Bioscience, Minneapolis, MN) were dissolved in soybean oil (Sigma-Aldrich) at different concentrations (0.2–5.0% w/v). Drugs were topically administered (5 μL) to cauterized corneas at 30, 60, and 120 min postcauterization. The CB₁R antagonist AM251 (N-(Piperidin-1-yl)-5-(4-iodophenyl)-1-(2,4-dichlorophenyl)-4-methyl-1H-pyrazole-3-carboxamide; Tocris Bioscience) was suspended in 10% dimethyl sulfoxide (DMSO, Sigma-Aldrich) and diluted in sterile saline. AM251 was injected at 2.0 mg/kg intraperitoneally (i.p.) fifteen min before cauterization. Capsaicin (1 μM, in 0.002% DMSO in sterile saline) was applied topically to eyes (5 μL) 6 h postinjury. The 5-HT_1A receptor antagonist WAY100635 (N-[2-[4-(2-Methoxyphenyl)-1-piperazinyl]ethyl]-N-2-pyridinylcyclohexanecarboxamide maleate; Tocris Bioscience) was dissolved in sterile saline and injected at 1.0 mg/kg i.p., 15 min before cauterization.

Data analysis

Individual animals in each treatment group were coded and experimental data were analyzed blinded. One-way analysis of variance (ANOVA) with Dunnett's or Tukey's multiple comparison post hoc tests was used, as appropriate, to compare data between experimental groups of three or more. t-Tests were used to compare two experimental groups. The number of animals in each group was 5–12. All data reported are represented as group mean±standard deviation. Data were considered significant at p<0.05.

Results

Corneal chemical injury results in hyperalgesia and inflammation

At 6 h, the pain score to 1 μM capsaicin was significantly increased in cauterized eyes (20±7, n=10) compared to sham control (11±4, n=6; p<0.01). At 12 h, no significant difference (p>0.05) was observed in pain score between sham control animals (7±3, n=5) compared to cauterized eyes (12±3, n=6). In addition, the pain score in cauterized eyes at 12 h was significantly lower than the pain score at 6 h (p<0.05; Fig. 1A).

FIG. 1.

Corneal chemical injury results in hyperalgesia and inflammation. (A) Pain responses to topical capsaicin challenge (1 μM) in noncauterized sham control eyes (n=5–6 per group) and cauterized eyes (n=6–10 per group) at 6 and 12 h postinjury. (B) Neutrophil expression in cauterized corneas at 6 and 12 h postinjury (n=5–6 per group). (C–E) Representative images of transverse sections of the central cornea from (C) sham control (noncauterized) corneas and cauterized corneas at (D) 6 h and (E) 12 h postinjury. Arrow in (E) points to one of many infiltrating neutrophils. Scale bar=50 μm. Values represent mean±SD. For statistical analysis, one-way ANOVA with Tukey's multiple comparison post hoc test was used. **p<0.01, *p<0.05. ANOVA, analysis of variance.

Immunohistochemical analysis was carried out to examine neutrophil migration, indicative of an inflammatory response, in the cornea 6 and 12 h after capsaicin challenge in either cauterized or sham eyes. Neutrophils were not observed in sham control corneas at 12 h (Fig. 1C). Figure 1B demonstrates, however, the presence of neutrophils at 6 and 12 h following cauterization (126±33, n=5, and 156±28, n=6, respectively; Fig. 1B, D, E).

Topical application of Δ⁸THC, CBD, and HU-308 reduces corneal pain and inflammation

Vehicle treatment produced an average pain score of 28±6 (n=8). Different doses of topical Δ⁸THC, CBD, and HU-308 were examined in WT mice to establish the effective drug concentrations required to reduce corneal pain compared to the vehicle-treated group (Fig. 2A). Administration of 0.5% and 1% Δ⁸THC produced a significant reduction in pain scores (18±6, n=6, p<0.05; 12±5, n=12, p<0.0001, respectively). Although at lower concentrations (0.2% and 0.4%) Δ⁸THC did not significantly affect the pain score (n=6 in each group; p>0.05). Topical application of 5% CBD also significantly reduced the pain score (15±3, n=10, p<0.001); however, 3% CBD was not effective at reducing the pain score (n=6, p>0.05). In addition, while 1% HU-308 did not produce a significant reduction in pain score (n=6, p>0.05), administration of 1.5% HU-308 was antinociceptive (17±4, n=6, p<0.01). Therefore, given their efficacy at reducing the pain response, 1% Δ⁸THC, 5% CBD, and 1.5% HU-308 were used for all further experiments.

FIG. 2.

Topical administration of Δ⁸THC, CBD, or HU-308 reduces corneal hyperalgesia and neutrophil infiltration in WT mice after corneal cauterization. (A) Dose-response for antinociceptive effects of Δ⁸THC (0.2–1.0%, n=6–12 per group), CBD (3% and 5%, n=6 and10, respectively), and HU-308 (1% and 1.5%, n=6 per group) following capsaicin challenge. (B) The number of neutrophils per section in corneas from WT mice treated with 1% Δ⁸THC, 5% CBD, or 1.5% HU-308 at 12 h postinjury compared to vehicle-treated eyes (n=6 per group). Values represent mean±SD. For statistical analysis, one-way ANOVA with Dunnett's post hoc test (compared to vehicle) was used. ****p<0.0001, ***p<0.001, **p<0.01, *p<0.05. Δ⁸THC, Δ⁸-tetrahydrocannabinol; CBD, cannabidiol; WT, wild-type.

Neutrophil infiltration into the cornea following treatment with cannabinoids was examined. Topical administration of 1% Δ⁸THC, 5% CBD, or 1.5% HU-308 significantly reduced neutrophil number (124±31, 144±16, and 73±22, respectively) compared to vehicle-treated eyes (205±21; p<0.0001, p<0.001, and p<0.0001, respectively; n=6 per group).

The antinociceptive and anti-inflammatory effects of Δ⁸THC, but not CBD, were mediated through CB₁R

Administration of the CB₁R antagonist AM251 (2.0 mg/kg, i.p.), before corneal cauterization and capsaicin stimulation, blocked the antinociceptive actions of Δ⁸THC (Fig. 3A; n=8, p>0.05), suggesting that Δ⁸THC acts via CB₁R to reduce corneal pain. However, the antinociceptive actions of 5% CBD were maintained in eyes pretreated with CB₁R antagonist AM251 (23±6, n=8), compared to vehicle-treated eyes plus AM251 (35±4, n=8, p<0.001; Fig. 3A).

FIG. 3.

The CB₁R antagonist AM251 reduces the antinociceptive and anti-inflammatory actions of Δ⁸THC but not CBD. (A) Pain score measured in WT mice at 6 h postcauterization and following administration of 5 μL of topical vehicle, 1% Δ⁸THC, or 5% CBD (n=8 per group) in mice preadministered with AM251 (2.0 mg/kg i.p.). (B) The number of neutrophils per section at 12 h postcauterization in corneas from WT mice preadministered with AM251 (2.0 mg/kg i.p.) and treated with 5 μL of vehicle, or either 1% Δ⁸THC or 5% CBD (n=6 per group). Values represent mean±SD. For statistical analysis, one-way ANOVA with Dunnett's post hoc test (compared to vehicle) was used. ***p<0.001. CB₁R, cannabinoid 1 receptor.

Likewise, the number of neutrophils in corneas from mice treated with AM251 and either 1% Δ⁸THC or vehicle was not significantly different (n=6, p>0.05). In contrast, 5% CBD treatment was still able to reduce neutrophils in corneas from mice treated with AM251 (95±44, n=6) versus vehicle-treated cauterized eyes from mice receiving AM251 (202±31, n=6, p<0.01; Fig. 3B).

The antinociceptive and anti-inflammatory effects of HU-308, but not Δ⁸THC or CBD, were mediated through CB₂R

The involvement of CB₂R in the antinocicpetive and anti-inflammatory effects of Δ⁸THC, CBD, and HU-308 was examined using CB₂R^−/− mice. Compared to WT mice receiving vehicle, the mean number of neutrophils in vehicle-treated CB₂R^−/− mice was significantly increased (mean difference 102±30, n=6 and 7, respectively, p<0.01); however, there was no significant difference in the pain score (n=8 in each group, p>0.05).

In CB₂R^−/− mice at 6 h postcauterization, compared to vehicle-treated eyes (27±7, n=8), application of 1% Δ⁸THC (8±5, n=12) or 5% CBD (19±4, n=7) significantly decreased the pain score (p<0.0001 and p<0.05, respectively). However, the antinociceptive effect of 1.5% HU-308 (n=7) was not significantly different compared to vehicle-treated animals (p>0.05; Fig. 4A), confirming the involvement of CB₂R in the antinociceptive effects of HU-308, but not THC or CBD.

FIG. 4.

The corneal antinociceptive and anti-inflammatory effects of HU-308, but not Δ⁸THC or CBD, are mediated via CB₂R. (A) Pain scores measured at 6 h postcorneal cauterization in CB₂R^−/− mice treated with 5 μL topical vehicle (n=8), or 1% Δ⁸THC (n=12), 5% CBD (n=7), or 1.5% HU-308 (n=7). (B) The number of neutrophils per section measured at 12 h postcauterization in corneas from CB₂R^−/− mice treated with 5 μL of vehicle (n=7), or 1% Δ⁸THC (n=6), 5% CBD (n=6), or 1.5% HU-308 (n=6). Values represent mean±SD. For statistical analysis, one-way ANOVA with Dunnett's post hoc test (compared to vehicle) was used. ****p<0.0001, **p<0.01, *p<0.05. CB₂R, cannabinoid 2 receptor.

Consistently, at 12 h postcauterization in CB₂R^−/− mice, the number of neutrophils in corneas receiving either 1% Δ⁸THC (123±50, n=6) or 5% CBD (187±28, n=6) was significantly less than vehicle-treated corneas (307±71, n=7; p<0.0001 and p<0.01, respectively). In HU-308-treated corneas (1.5%) from CB₂R^−/− mice (n=6), there was no significant difference in neutrophil numbers compared to vehicle-treated eyes (p>0.05; Fig. 4B).

CBD acts at 5-HT_1A receptors to reduce corneal pain and inflammation

The corneal antinociceptive and anti-inflammatory effects of CBD were independent of CB₁R or CB₂R. Therefore, we examined an alternative non-cannabinoid receptor, 5-HT_1A, which has been reported as a target for CBD in other tissues.²⁹ Treatment of mice with the 5-HT_1A receptor antagonist WAY100635 (1.0 mg/kg i.p.) was able to completely eliminate the antinociceptive actions of CBD in cauterized cornea (Fig. 5A; n=8 in each group, p>0.05). In mice treated with WAY100635, the reduction in corneal neutrophils in cauterized eyes seen with CBD treatment was also blocked (n=6 in each group, p>0.05), suggesting that 5-HT_1A is the target receptor for CBD-mediated antinociceptive and anti-inflammatory actions in the cornea (Fig. 5B).

FIG. 5.

The corneal antinociceptive and anti-inflammatory effects of CBD are mediated through 5-HT_1A receptor. (A) Pain score measured at 6 h postcauterization in WT mice preadministered with the 5-HT_1A receptor antagonist, WAY100635 (1.0 mg/kg i.p.), and treated topically with 5 μL of either vehicle (n=8) or 5% CBD (n=8). (B) The number of neutrophils per section measured at 12 h postcauterization in corneas from WT mice treated with 5 μL of either vehicle (n=6) or 5% CBD (n=6). Values represent mean±SD. For statistical analysis, unpaired t-tests were used.

Discussion

Our results provide novel evidence that the phytocannabinoids Δ⁸THC and CBD and synthetic cannabinoid derivative HU-308 are antinociceptive and anti-inflammatory in an experimental model of corneal hyperalgesia. Furthermore, we demonstrate that the actions of these cannabinoids are mediated via distinct receptor targets that include CB₁R and CB₂R, as well as 5-HT_1A receptor.

In the mammalian cornea, expression of CB₁R has been reported to colocalize with TRPV1,³⁴ the latter of which is expressed in corneal epithelium³⁵ and endothelium,³⁶ and sensory nerve endings of the ophthalmic branch of the trigeminal nerve innervating the cornea.³⁷ TRPV1 is activated following damage to cornea, culminating in activation of corneal nerves and local inflammation. The release of proinflammatory cytokines and neuropeptides, including nerve growth factor and substance P, contributes to neurogenic inflammation and can lead to corneal nerve sensitization.^34,38 In sensory neurons isolated from rat dorsal root ganglia, activation of CB₁R by the cannabinoid agonist ACEA (arachidonoyl-2′- chloroethylamide) prevented nerve growth factor-induced sensitization of the TRPV1 receptor. This action was blocked by the CB₁R antagonist AM251, suggesting that the activation of CB₁R may produce analgesia by desensitization of TRPV1 receptors.³⁹

Our results demonstrating that topical Δ⁸THC, acting at CB₁R, reduces hyperalgesia following corneal injury are in line with these findings. In addition, we also demonstrated that Δ⁸THC was able to reduce the neutrophil recruitment to the cornea observed at later time points following corneal epithelial damage. The inhibition of neutrophil recruitment was blocked with the treatment of CB₁R antagonist AM251, but was still present in CB₂R^−/− mice. This suggests that the activation of CB₁R by Δ⁸THC is important in mitigating the innate immune response following corneal injury, which may contribute to corneal nerve sensitization.

The importance of peripheral CB₁R in our study is also consistent with the actions of Δ⁹THC reported in other models of both acute inflammatory and neuropathic pain. For example, administration of Δ⁹THC (1 mg/kg i.p.) in a rat model of acute muscle pain produced antinociceptive effects, which was blocked by the CB₁R antagonist AM281 and to a lesser extent by the CB₂R antagonist AM630 (0.5 mg/kg i.p.).⁴⁰ Furthermore, in a model of inflammatory and neuropathic pain, mice lacking CB₁R in peripheral nociceptive neurons showed a reduced analgesic effect to local and systemic administration of the cannabinoid WIN55,212-2. With intrathecal application, the analgesic effect of WIN55,212-2 was absent, suggesting that peripheral CB₁R in nociceptive neurons plays an important role in producing the analgesic effects of cannabinoids.⁴¹

Our data implicate 5-HT_1A receptors, and not cannabinoid receptors, in both antinociceptive and anti-inflammatory actions of CBD in an experimental model of corneal injury. We showed that the actions of CBD were completely blocked by the 5-HT_1A antagonist WAY100635, but were still present after CB₁R block or in CB₂R^−/− mice. CBD has been reported in other in vitro and in vivo models to bind to 5-HT_1A receptors.^27,29,42 Using a heterologous cell expression system, Russo et al. reported that CBD bound to both human and rat 5-HT_1A receptors with micromolar affinity, and displaced the agonist [³H]8-OH-DPAT in a concentration-dependent manner.²⁹ In addition, CBD increased [³⁵S]GTPγS binding, and decreased forskolin-stimulated cAMP production, which was blocked by the specific 5-HT_1A antagonist NAN-190.²⁹

In line with our findings of CBD activity at the 5-HT_1A receptor, a study by Ward et al. reported that CBD administration could prevent chemotherapy-induced neuropathic pain associated with paclitaxel treatment.⁴³ In this study, CBD was administered chronically for 14 days and prevented the onset of paclitaxel-induced mechanical and thermal sensitivity in female mice. A subsequent report showed that a subchronic dosing regimen of 2.5–10 mg/kg CBD (i.p.) was also effective in preventing paclitaxel-induced mechanical sensitivity. This effect was blocked by a 5-HT_1A antagonist (WAY100635), but not a CB₁R (SR141716) or CB₂R antagonist (SR144528), further supporting the role of 5-HT_1A in mediating the actions of CBD in preventing neuropathic pain.⁴³

HU-308 has been reported as a selective CB₂R agonist.³⁰ In our model of corneal hyperalgesia, the antinociceptive and anti-inflammatory actions of HU-308, unlike Δ⁸THC and CBD, were absent in CB₂R^−/− mice, validating target specificity for this cannabinoid at CB₂R. This is the first time a CB₂R agonist has been demonstrated to reduce corneal pain, although CB₂R activation has been reported to reduce ocular inflammation.^20,44 In experimental uveitis, Toguri et al.¹⁹ reported that CB₂R activation reduced leukocyte/endothelial adhesion in the iridial microvasculature as well as inhibited release of proinflammatory mediators, including TNFα, IL1β, Il6, CCL5, and CXCL2. Conversely, a CB₂R antagonist, AM630, increased leukocyte/endothelial adhesion in experimental uveitis,¹⁹ suggesting that CB₂R activity in the eye is immunosuppressive during inflammation. In a mouse model of proliferative vitreoretinopathy, CB₂R^−/− or pharmacological block of CB₂R, also produced increased inflammation and a more severe pathology.⁴⁴ Another study, in a mouse model of endotoxemia, has shown increased neutrophil recruitment to the spleen in CB₂R^−/− mice compared to WT control.⁴⁵ In line with these results, in our experiments we observed an increase in the mean number of neutrophils in cauterized corneas in CB₂R^−/− mice, suggesting that loss of constitutive CB₂R activity is proinflammatory in ocular tissues.^20,44

Reports of the antinociceptive and antiallodynic efficacy of CB₂R agonists have also been reported in other experimental models of hyperalgesia and chronic inflammatory neuropathic pain.⁴⁶ While our study in cornea used a relatively acute dosing regimen, the utility of CB₂R agonists used chronically was previously reported in a mouse model of paclitaxel-induced neuropathic pain.⁴⁷ The authors reported that chronic CB₂R activation with the CB₂R-preferring agonist AM1710 was able to reverse paclitaxel-induced allodynia, an effect that was blocked in WT mice treated with the CB₂R antagonist AM630, or in CB₂R^−/− mice. In comparison to repeated dosing with agonists such as THC that produced behavioral effects and tolerance via CB₁R activation, no similar effects were observed with the CB₂R-preferring agonist AM1710. Furthermore, using intrathecal cannabinoid administration, this study identified a possible role for spinal CB₂R in the antiallodynic actions of AM1710, as well as a reduction in proinflammatory cytokines, in paclitaxel-treated mice.⁴⁷ Increased CB₂R expression has also been reported in human peripheral nerves after injury, and CB₂R agonist-mediated inhibition of capsaicin responses was observed in cultured human dorsal root ganglion sensory neurons.⁴⁸ Our data demonstrating the antinociceptive and anti-inflammatory actions of CB₂R activation in cornea, together with these studies, further support the utility of CB₂R agonists for treating inflammatory pain.

Conclusion

Our study showed that topical application of the phytocannabinoids Δ⁸THC and CBD, and the cannabinoid derivative HU-308, reduced corneal hyperalgesia and neutrophil infiltration resulting from superficial chemical injury of corneal epithelium. The effects of these cannabinoids were mediated by distinct receptors, including CB₁R and CB₂R, as well as 5-HT_1A receptors. This suggests that when used either as sole agents or in combination, these cannabinoids could be effective agents in the treatment of ocular pain and inflammation resulting from corneal surface injuries.

Abbreviations Used

Δ⁸THC

Δ⁸-tetrahydrocannabinol

Δ⁹THC

Δ⁹-tetrahydrocannabinol

CBD

cannabidiol

CB₁R

cannabinoid 1 receptor

CB₂R

cannabinoid 2 receptor

CNP

corneal neuropathic pain

DMSO

dimethyl sulfoxide

ECS

endocannabinoid system

PBS

phosphate-buffered saline

TRPV1

transient receptor potential cation channel subfamily V member 1

WT

wild-type

Acknowledgments

This research was supported by the Canadian Institutes of Health Research (MOP-97768). D.T. was supported, in part, through the NSGS. The authors thank Janette Nason and Anjali Ghimire for their technical assistance.

Author Disclosure Statement

M.E.M.K. is the founder and director of Panag Pharma, Inc. Panag develops phytotherapeutics for local and regional treatment of pain and inflammation. D.T., E.A.C., J.T.T., A.-M.S., and M.D.C. have no existing competing financial interests.